Population

Median Income

Percent White

Notes

• Between 2010 and 2017, Manhattan experienced a population growth from 1.58 million to 1.65 million.
• The median income of the city as whole went from 69k to 86k.
• The percent white stayed about the same, at around 56%-57.

Top Complaints in 2010

311 Calls Over the Year

Top Complaints in 2011

311 Calls Over the Year

Top Complaints in 2012

311 Calls Over the Year

Top Complaints in 2013

311 Calls Over the Year

Top Complaints in 2014

311 Calls Over the Year

Top Complaints in 2015

311 Calls Over the Year

Top Complaints in 2016

311 Calls Over the Year

Top Complaints in 2017

311 Calls Over the Year

Total 311 Calls

With Change in Median Income

With Change in Percent White

Notes

• The number of 311 Calls increased steadily between 2012 and 2016, with a peak in 2016.
• Median income: There were many tracts that experienced between a 0% and 100% increase in median income between 2010 and 2017. For these tracts, there wasn’t much change in the number of 311 calls per person across the income changes. There seems to be a slight trend upward for tracts experiencing a large median income increase (like > 200%), but there are not enough data points in this area to make any significant conclusions. The majority of tracts had an increase in the number of 311 calls per person, regardless of percent income change. The percent change of the median household income does not seem to be a good indicator of a trend in 311 calls.
• Most tracts had demographic shifts, either by and increase or decrease in percent white by roughly 10%. For tracts that experienced this from 2010 to 2017, there was no signficant change in the number of 311 calls per person. In fact, most of these tracts sit around 0 in change in calls per person. Fpr tracts that have a change less than -10% or greater than +10%, there seems to be a pull upwards and downwards, respectively, but there do not seem to be enough data points to make a signficant conclusion.

From looking at these data, I found there were no conclusive general trends with the sheer number of calls per person for changes in either income or percent white. Earlier we saw that most 311 complaints have to do with either noise or heating, so I decided to dig a bit deeper in these specific call types. I was interested to see if there were any trends with noise or heat complaints and demographic shifts. To do this, I started by first looking at just single years, and then at the change between 2010 and 2017.

So far, my answer to my original question (As neighborhoods gentrify, do the amount of 311 calls increase?) is no, using my rudimentary measures of gentrification and the total number of 311 calls.

2010 Percent White

2010 Median Income

2017 Percent White

2017 Median Income

Change Percent White

Change Median Income

Notes

I took a look at the number of heat calls per person in each tract and plotted them against both median income and percent white of the tract. I did so for both 2010 and 2017.

For percent white in 2010 and 2017: For both of these years, you can see that tracts with a lower percent white people are more likely to call 311 for a heating complaint. Communities that are less than 50% white have a rate of 0.05 heating calls per person, while those greater than 50% white have about a 0.025 rate of heating calls per person. The shift from more heating complaints to fewer heating complaints occurs about when a tract becomes 50% white or greater.

For median income in 2010 and 2017: A similar pattern as the percent white trend appears for plots against median income of a tract. Poorer tracts are more likely to call a 311 heat complaint. Specifically, tracts that have a median income of less than $75k experience a rate of around 0.05 heating calls per person, while tracts over this threshold see a rate of around 0.025 heating calls per person.

311 heating calls address issues dealing with the heat of a building, or the hot water in a building, as described by the City of New York. A hypothesis: For tracts with lower median incomes, the buildings people live in might be in a poorer condition than wealthy tracts. This might coincide with worse heating or water systems, thus resulting in more heating complaints. The percent whiteness of a tract might have some relationship to the median income of a tract, given historical socioeconomic trends.

Now let’s take a look at the change from 2010 to 2017. As we saw in the first tab, most tracts experienced a change in +/-10% in percent white. For these tracts, there was no significant change in the number of heating complaints made. For the tracts that had a median income increase from 0% to 100% also had no signficant change in the number of heating complaints. (There is, however, a trend upwards as the household income increases by more than 100%, but there are too few points there to make any signficant conclusions). Perhaps this is because buildings don’t change much in the span of 7 years, so buildings with heating problems in 2010 would continue to have the same heating problems in 2017, regardless of who’s living there.

So in total, my takeaways: race and income somewhat determine the number of heating complaints in a tract, but the change in tract demographics between 2010 and 2017 does not seem to affect the heating complaints. The answer to my initial question, therefore, continues to be a no.

2010 Percent White

2010 Median Income

2017 Percent White

2017 Median Income

Change Percent White

Change Median Income

Takeaways

I took a look at the number of noise complaint calls per person in each tract and plotted them against both median income and percent white of the tract. I did so for both 2010 and 2017.

For percent white in 2010 and 2017: The plots show that the number of noise complaints peaks at a certain percentage white of a tract. In 2010, a tract that was %50 white was more likely to call 311 for a noise complaint than a tract that was 30% white or 70% white. However, the rate of noise calls per person starts to pick up again at around 90% white. In 2017, there’s a similar trend, but with the peak in noise complaint rate to be at around 65% white. For communities either less white or more white than this, the rate of calling noise complaints was smaller. Especially for tracts that are greater than 70% in 2017, there were very few noise complaints called. While this observation doesn’t directly answer my question about gentrification and 311 call trends, there’s something interesting to note: the most noise complaints come from tracts where there is a mix of people of different races living together. Gentrified neighborhoods could very well be among those tracts that have a mix of white people and non-white people, where there is a peak in the number of noise complaints made.

For median income in 2010 and 2017: In 2010, there pretty consistent noise complaint rates among tracts of different income levels, with a slight dip at around $75k. For tracts with either smaller or larger median incomes, the noise complaint rate increases. In 2017, we see a similar trend, with a pretty consistent noise complaint rate across income levels, except for a dip around 85k. We can’t conclude anything too significant, but I suppose we could say that there are slightly fewer noise complaint calls being made in tracts that are middle-income, around 75k to 85k.

When looking at the change between 2010 and 2017, it’s clear to see that in general, the number of noise complaints has increased dramatically (around 0.1 call per person more). For those same tracts that we talked about earlier, that changed between -10% white to +10% white, there was a consistent increase in noise complaints, but not skewed to one side. This visualzation indicates that neighborhoods that change to be way less white experience more noise complaints, but there don’t seem to be enough data around there to make a definitive conclusion. The same goes for change in median income: while noise complaints increased overall, the tracts that changed from a 0% to 100% increase in median income had roughly the same increase in noise complaint rate.

Again, we saw some interesting trends in terms of demographics of a tract for a single year and noise complaints, but there’s no signficant conclusion about neighborhoods than change over time.